Battery management system and operating method thereof

Abstract

A battery management system includes: a contactor connected to a battery; a battery manager configured to output a control signal for controlling a state of the contactor to the contactor; a signal processor configured to receive a radio frequency (RF) signal generated during state switching of the contactor, and generate a plurality of signal strength values corresponding to a plurality of frequency bands, respectively, based on the RF signal; and a contactor state estimator configured to generate a state estimation value for estimating a state of the contactor by receiving the plurality of signal strength values and inputting the plurality of signal strength values into a pre-trained neural network model.

Description

[0001] Cross - reference to related applications

[0002] This application claims the priority and benefit of Korean Patent Application No. 10 - 2023 - 0165832, filed with the Korean Intellectual Property Office on November 24, 2023, the entire disclosure of which is incorporated herein by reference. Technical field

[0003] Aspects of one or more embodiments relate to a battery management system and a method of operating the same. Background art

[0004] A battery management system (BMS) for a vehicle manages the performance and safety of a battery pack in an electric vehicle (xEV). The BMS monitors the state of charge, voltage, temperature, etc. of the battery and controls the charging and discharging of the battery based on these. The BMS also detects abnormal states of the battery and can take protective measures if necessary.

[0005] A contactor is a component that can be utilized in a BMS and controls the electrical connection between the battery and the vehicle's electrical system. The contactor switches between an open state and a closed state to control the charging and discharging of the battery. A noise signal may be generated during the process of switching the state of the contactor, and the noise signal can be used to detect the normal state and abnormal state of the contactor.

[0006] In some methods, the state of the contactor can be detected by using a physical sensor. The physical sensor can determine the state of the contactor by measuring the position, current, voltage, etc. of the contactor. However, in some methods, additional hardware may be utilized, and due to errors or malfunctions of the physical sensor, the reliability of the result of determining the state of the contactor may be reduced.

[0007] The above information disclosed in this background section is only for enhancing the understanding of the background, and thus the information discussed in this background section does not necessarily constitute prior art. Summary of the invention

[0008] Aspects of one or more embodiments include a battery management system and a method as follows: detecting a radio frequency (RF) signal generated during a state transition of a contactor in the battery management system, and determining the state of the contactor based on a signal strength value generated from the RF signal and a pre - trained neural network model.

[0009] However, the features according to the embodiments of the present disclosure are not limited to the above - mentioned features, and other features not mentioned can be clearly understood by those of ordinary skill in the art from the description of the embodiments according to the present disclosure below.

[0010] Other aspects will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of the disclosure.

[0011] According to one or more embodiments, a battery management system includes a contactor connected to a battery, a battery manager configured to output a control signal for controlling a state of the contactor to the contactor, a signal processor configured to receive a radio frequency (RF) signal generated during switching of the state of the contactor and generate a plurality of signal strength values respectively corresponding to a plurality of frequency bands based on the RF signal, and a contactor state estimator configured to generate a state estimation value for estimating the state of the contactor by receiving the plurality of signal strength values and inputting the plurality of signal strength values into a pre-trained neural network model.

[0012] According to some embodiments, the battery manager may further be configured to output a trigger signal to the signal processor in response to output of the control signal, and the signal processor may further be configured to capture the RF signal in response to the trigger signal.

[0013] According to some embodiments, the signal processor may include an antenna configured to receive the RF signal, an RF tuner configured to output a band signal in a preset frequency band in the RF signal, an analog-to-digital converter (ADC) configured to convert the band signal into a digital signal, a digital signal processor (DSP) configured to convert the digital signal into a frequency domain signal by performing a Fourier transform on the digital signal and generate a plurality of signal strength values from the frequency domain signal, and a signal processor controller configured to control the antenna, the RF tuner, the analog-to-digital converter (ADC), and the digital signal processor (DSP) in response to the trigger signal and send the plurality of signal strength values to the contactor state estimator.

[0014] According to some embodiments, the preset frequency band may be above 100 kHz and below 5 MHz.

[0015] According to some embodiments, the widths of the plurality of frequency bands may be the same as each other on a logarithmic scale.

[0016] According to some embodiments, the widths of the plurality of frequency bands may be the same as each other.

[0017] According to some embodiments, the neural network model may include a convolutional neural network (CNN) model, and may be pre-trained using training data including a plurality of signal strength values generated by preprocessing the state of the contactor switched by the control signal of the battery manager and the RF signal generated during switching of the state of the contactor.

[0018] According to some embodiments, the control signal may include a turn-on signal for switching the contactor to an open state and a turn-off signal for switching the contactor to a closed state.

[0019] According to some embodiments, the pre-trained neural network model may be configured to output a state estimate value based on a plurality of signal strength values received as input and a confidence score of the state estimate value.

[0020] According to some embodiments, the battery manager may be configured to detect whether the contactor is in an abnormal state based on the state estimate value and the confidence score of the state estimate value.

[0021] According to some embodiments, the battery manager may also be configured to determine the state of the contactor as an abnormal state based on at least one of a closed state of the state estimate value of the contactor in response to the turn-on signal, an open state of the state estimate value of the contactor in response to the turn-off signal, and a welded state of the state estimate value of the contactor.

[0022] According to one or more embodiments, a method of operating a battery management system of a contactor connected to a battery includes: outputting a control signal for controlling the state of the contactor, receiving an RF signal generated during switching the state of the contactor, generating a plurality of signal strength values respectively corresponding to a plurality of frequency bands based on the RF signal, and generating a state estimate value for estimating the state of the contactor by inputting the plurality of signal strength values into a pre-trained neural network model.

[0023] According to some embodiments, the method may further include extracting a trigger signal in response to the output of the control signal and capturing the RF signal in response to the trigger signal.

[0024] According to some embodiments, the method may further include receiving the RF signal, outputting a band signal in a preset frequency band from the RF signal, converting the band signal into a digital signal, transforming the digital signal into a frequency domain signal by performing a Fourier transform on the digital signal and generating a plurality of signal strength values from the frequency domain signal, and extracting the plurality of signal strength values in response to the trigger signal.

[0025] According to some embodiments, the control signal may include a turn-on signal for switching the contactor to an open state and a turn-off signal for switching the contactor to a closed state.

[0026] According to some embodiments, the pre-trained neural network model may be configured to generate a state estimate value based on a plurality of signal strength values received as input and a confidence score of the state estimate value.

[0027] According to some embodiments, the method may further include detecting whether the contactor is in an abnormal state based on the state estimation value and the confidence score of the state estimation value.

[0028] According to some embodiments, the method may further include determining the state of the contactor as an abnormal state based on at least one of a closed state of the state estimation value of the contactor in response to the on signal, an open state of the state estimation value of the contactor in response to the off signal, and a welded state of the state estimation value of the contactor.

[0029] According to one or more embodiments, a computer program is stored in a medium to perform the above method of operating a battery management system by using a computing device. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above and other aspects, features, and characteristics of certain embodiments of the present disclosure will become more apparent from the following description taken in conjunction with the accompanying drawings, in which:

[0031] The following drawings attached to the specification illustrate embodiments of the present disclosure and, together with the detailed description of the present disclosure described below, are used to further understand the technical spirit of the present disclosure. Therefore, the present disclosure should not be construed as limited to the content described in the drawings.

[0032] Figure 1 Schematically shows a battery management system according to some embodiments of the present disclosure;

[0033] Figure 2 Schematically shows a process of operating a battery management system according to some embodiments of the present disclosure;

[0034] Figure 3 Is a schematic block diagram of a computing device for performing a method of operating a battery management system according to some embodiments of the present disclosure;

[0035] Figure 4 Is a flowchart showing a method of operating a battery management system according to some embodiments of the present disclosure;

[0036] Figure 5A Is a view showing an example structure of a pre-trained artificial neural network (ANN) model according to some embodiments of the present disclosure;

[0037] Figure 5B Schematically shows an example of a pre-trained ANN model according to some embodiments of the present disclosure; and

[0038] Figure 5C Is an example schematically showing a process of training a pre-trained ANN model according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0039] Reference will now be made in more detail to aspects of some embodiments, examples of which are illustrated in the accompanying drawings, wherein like reference numerals always refer to like elements. In this regard, the present embodiments may have different forms and should not be construed as limited to the descriptions set forth herein. Accordingly, the embodiments are described below only by reference to the accompanying drawings to explain aspects of the present specification. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items. Expressions such as "at least one of..." modify the entire list of elements when preceding the list of elements, rather than modifying a single element in the list.

[0040] Hereinafter, example embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Before the description, terms or words used in the specification and claims should not be construed only in their normal or dictionary meanings, but should be interpreted as meanings and concepts consistent with the technical spirit of the present disclosure based on the principle that the inventor can appropriately define terms to best describe the present disclosure. Accordingly, it should be understood that the embodiments described in the specification and the components shown in the drawings are only some embodiments of the present disclosure and do not represent all the technical spirits of the present disclosure. Thus, there may be examples of various equivalents and modifications that can replace the embodiments and components at the time of filing this application. It will be further understood that when used herein, the terms "comprises", "comprising", "includes" and / or "including" specify the presence of the stated features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. When describing embodiments of the present disclosure, "~can" and "~may" may include "one or more embodiments of the present disclosure".

[0041] Although terms such as first, second, etc. may be used herein to describe various elements or components, these elements or components should not be limited by these terms. These terms are only used to distinguish one element or component from another, and the first element may be referred to as the second element unless the context clearly indicates otherwise.

[0042] Unless the context clearly indicates otherwise, each component may be singular or plural throughout the specification.

[0043] Throughout the specification, "A and / or B" means A, B, or A and B, unless the context clearly indicates otherwise. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items. Unless the context clearly indicates otherwise, "C or D" means C or more and D or less.

[0044] Figure 1 Schematically shows a battery management system (BMS) according to some embodiments of the present disclosure.

[0045] Reference Figure 1 , BMS1000 may include a contactor (or switch) 11 connected to the battery, a battery manager 100, a signal processor 200, and a contactor state estimator 300.

[0046] BMS1000 may be an important component for managing the performance and safety of a battery pack in an electric vehicle (xEV). BMS1000 may monitor the state of charge, voltage, temperature, etc. of the battery and control the charging and discharging of the battery based on these.

[0047] The contactor 11 may be one of the components that play an important role in BMS1000. For example, the contactor 11 may control the electrical connection between the battery and the vehicle's electrical system. The contactor 11 may switch its state to control the charging and discharging of the battery. For example, the state of the contactor 11 may be switched to an open state or a closed state. During the process of switching the state of the contactor 11, a radio frequency (RF) signal SIG RF . RF signal SIG RF may include noise, and the RF signal SIG RF including noise can be used to detect the normal state and abnormal state of the contactor 11.

[0048] The battery manager (or battery management component or battery management circuit) 100 may be a component that controls the operation of the contactor 11 and outputs signals for diagnosing the state of the contactor 11 to the signal processor 200 and the contactor state estimator 300. The battery manager 100 may output a control signal SIG CTR . Control signal SIG CTR may include an on signal for switching the contactor 11 to an open state and an off signal for switching the contactor 11 to a closed state. The battery manager 100 may output a trigger signal SIG CTR to the signal processor 200 and the contactor state estimator 300 in response to the output of the control signal SIG TRG . For example, the trigger signal SIG TRG may be a signal including a command for detecting the RF signal SIG RF generated from the contactor 11 and performing sampling. The battery manager 100 may include a microcontroller unit (MCU). According to some embodiments, the battery manager 100 may include a processor and a memory, where the memory stores instructions that, when executed by the processor, cause the battery manager 100 to perform various operations as described herein.

[0049] The battery manager 100 can detect whether the contactor 11 is in an abnormal state based on the estimated value of the state of the contactor 11 and the confidence level of the estimated value of the state. For example, the state of the contactor 11 according to the estimated value of the state of the contactor 11 can be at least one of a closed state, an open state, or a welded state. If the state of the contactor 11 is a welded state, the battery manager 100 can determine the state of the contactor 11 as an abnormal state.

[0050] The signal processor 200 can be a software-defined radio (SDR). An SDR can be a wireless communication system implemented by software rather than components previously implemented as analog hardware (e.g., mixers, filters, amplifiers, modulators / demodulators, detectors, etc.). If an SDR is used, a single piece of hardware can capture, demodulate, and access RF signals with a wide radio frequency. For example, an SDR can capture a wide spectrum and selectively analyze a specific part of the RF signal. The signal processor 200 (e.g., an SDR) can capture the RF signal SIG generated during the switching of the state of the contactor 11 in the frequency domain RF 。The RF signal SIG RF can include a noise pattern, and the noise pattern can reflect the operating state of the contactor 11. In some embodiments, if the contactor 11 is in an open, closed, or welded state, the noise pattern can have a unique pattern. The signal processor 200 can generate multiple signal strength values SS corresponding to multiple frequency bands based on the RF signal SIG RF respectively.

[0051] The signal processor 200 can capture the RF signal SIG generated from the contactor 11 in response to the trigger signal SIG TRG 。If the battery manager 100 outputs a control signal SIG for controlling the state of the contactor 11 to the contactor 11 RF ,the signal processor 200 can receive the trigger signal SIG from the battery manager 100 CTR 。Refer to TRG 。The components included in the signal processor 200 and their operation processes are described in more detail. Figure 2 。

[0052] The contactor state estimator 300 can generate an estimated value for estimating the state of the contactor 11 by receiving multiple signal strength values SS and inputting the multiple signal strength values SS into a pre-trained neural network model. According to some embodiments, the neural network model can be a convolutional neural network (CNN) model. According to some embodiments of the present disclosure, the neural network model can be generated by a processor for a specific purpose and can refer to an artificial neural network (ANN) trained by machine learning or deep learning techniques. Refer to the following Figures 5A to 5CDescribe aspects of an example structure of the neural network described above in more detail. The contactor state estimator 300 may include an MCU. The contactor state estimator 300 may also include a noise pattern processing unit (or noise pattern processor). According to some embodiments, the contactor state estimator 300 may include a processor and a memory, where the memory stores instructions that, when executed by the processor, cause the contactor state estimator 300 to perform various operations as described herein.

[0053] According to some embodiments, a neural network model may be pre-trained using training data including a plurality of signal strength values SS generated by preprocessing the state of the contactor 11 switched by the control signal SIG of the battery manager 100 and the RF signal SIG generated during the switching of the state of the contactor 11. The neural network model may output a state estimation value SEV based on the confidence scores of the plurality of signal strength values SS and the state estimation value SEV received as inputs. The contactor state estimator 300 may send the state estimation value SEV of the contactor 11 and the confidence score data of the state estimation value SEV to the battery manager 100. CTR switched and the RF signal SIG generated during the switching of the state of the contactor 11 RF are generated. The neural network model may output a state estimation value SEV based on the confidence scores of the plurality of signal strength values SS and the state estimation value SEV received as inputs. The contactor state estimator 300 may send the state estimation value SEV of the contactor 11 and the confidence score data of the state estimation value SEV to the battery manager 100.

[0054] Figure 2 Aspects of the BMS1000 for implementing the process of operating the BMS according to some embodiments of the present disclosure are schematically illustrated.

[0055] Referring Figure 2 to, BMS1000 may diagnose the state of the contactor 11 by identifying the noise pattern of the frequency-based RF signal SIG generated during the switching of the state of the contactor 11 and analyzing the noise pattern of the RF signal SIG using a pre-trained ANN. RF of the noise pattern and by analyzing the noise pattern of the RF signal SIG using a pre-trained ANN RF to diagnose the state of the contactor 11.

[0056] The BMS1000 may include a contactor 11 connected to the battery, a battery manager 100, a signal processor 200, and a contactor state estimator 300.

[0057] The contactor 11 may be one of the components that play an important role in the BMS1000. For example, the state of the contactor 11 may be an open state or a closed state. The RF signal SIG RF may be generated from the contactor 11 during the process of switching the state of the contactor 11.

[0058] The battery manager 100 may output a control signal SIG for controlling the state of the contactor 11 to the contactor 11 CTR . The control signal SIG CTRIt may include an on - signal for switching the contactor 11 to the off state and an off - signal for switching the contactor 11 to the on state. The battery manager 100 may output a trigger signal SIG CTR in response to the output of the control signal SIG TRG to the signal processor 200. The battery manager 100 may output the trigger signal SIG TRG to the contactor state estimator 300.

[0059] The battery manager 100 may detect whether the contactor 11 is in an abnormal state based on the state estimate value SEV of the contactor 11 and the confidence level of the state estimate value SEV. If the state of the contactor 11 according to the state estimate value SEV of the contactor 11 in response to the on - signal is the on state, the battery manager 100 may determine the state of the contactor 11 as an abnormal state. If the state of the contactor 11 according to the state estimate value SEV of the contactor 11 in response to the off - signal is the off state, the battery manager 100 may determine that the state of the contactor 11 is in an abnormal state. If the state of the contactor 11 according to the state estimate value SEV of the contactor 11 is in a welded state, the battery manager 100 may determine the state of the contactor 11 as being in an abnormal state.

[0060] The battery manager 100 may include an MCU. The MCU of the battery manager 100 may send an operation command to the contactor 11. For example, the MCU may send an operation command to the power system to activate the contactor 11. For example, the operation command may be a command indicating contactor on or a specific voltage / current level.

[0061] The MCU of the battery manager 100 may send a trigger signal SIG TRG to the signal processor 200, and this trigger signal SIG TRG commands to start detecting the RF signal SIG RF of the contactor 11. For example, the signal processor 200 may include a noise pattern processing unit. For example, the trigger signal SIG TRG may be a bytecode such as START_NOISE_DETECTION (start noise detection). The MCU of the battery manager 100 may send an operation command to the contactor 11 and, at the same time, send the trigger signal SIG TRG to the signal processor 200. For example, the trigger signal SIG TRG may include a timestamp, and thus, the detection of the RF signal SIG RF may be synchronized with the trigger signal SIG TRG . For example, the trigger signal SIG TRGIt can be sent through the Serial Peripheral Interface (SPI) protocol. However, embodiments according to the present disclosure are not limited thereto, and any communication protocol for sending data or commands in the BMS can be applied.

[0062] The MCU of the BMS1000 can adjust system parameters based on the analysis data according to the state estimation value SEV of the contactor 11. For example, the MCU can diagnose the state of the contactor 11 based on the analysis data, and according to some embodiments, determine protection measures for the BMS1000. For example, the protection measures can include setting a maximum noise amplitude threshold, adjusting the operating voltage of the contactor 11, etc. However, the above description is an example, and the above description does not limit the embodiments according to the present disclosure.

[0063] The signal processor 200 can include an antenna 210, an RF tuner 220, an analog-to-digital converter (ADC) 230, a digital signal processor (DSP) 240, and a signal processor controller 250. The signal processor 200 can be an SDR. Hereinafter, in Figure 2 the embodiments of, the signal processor 200 is described as an SDR. In response to the trigger signal SIG TRG received from the battery manager 100, the signal processor 200 can capture the RF signal SIG RF generated during the switching of the state of the contactor 11 in the frequency domain. TRG The signal processor 200 can receive the trigger signal SIG RF from the MCU, and simultaneously (or concurrently) start sampling the RF signal SIG RF received from the contactor 11. For example, the signal processor 200 can sample the RF signal SIG RF including the noise signal. The signal processor 200 can extract a plurality of signal strength values SS by sampling the RF signal SIG RF and generate a state estimation value SEV of the contactor 11 based on the plurality of signal strength values SS. For example, the signal processor 200 can generate noise data such as an analog voltage value by sampling the RF signal SIG

[0064] including noise. Hereinafter, the components included in the signal processor 200 are described in more detail. RF The antenna 210 can receive the RF signal SIG RF generated during the switching of the state of the contactor 11 through the environment. RF The antenna 210 can convert the RF signal SIG into an electrical signal and send the electrical signal to the RF tuner 220. For example, the antenna 210 can receive a broadband frequency signal with a frequency above 1 kHz and below 1 GHz.

[0065] The RF tuner 220 can output a band signal SIG in a preset frequency band from the RF signal SIG RF For example, the RF tuner 220 can receive the RF signal SIG from the antenna 210 BD and down-convert the frequency of the RF signal SIG RF to a preset frequency band. According to some embodiments, the preset frequency band can be above 100 kHz and below 5 MHz. For example, the RF tuner 220 can output a band signal SIG having a frequency above 100 kHz and below 5 MHz from an RF signal SIG having a frequency above 1 kHz and below 1 GHz RF For example, the RF tuner 220 can output a band signal SIG having a frequency above 100 kHz and below 5 MHz from an RF signal SIG having a frequency above 1 kHz and below 1 GHz RF According to some embodiments, the widths of multiple frequency bands can be the same as each other on a logarithmic scale. According to other embodiments, the widths of multiple frequency bands can be the same as each other. The RF tuner 220 can send the down-converted band signal SIG BD to the ADC 230. BD to the ADC 230.

[0066] The ADC 230 can convert the band signal SIG BD into a digital signal SIG DIG The ADC 230 can sample the band signal SIG in analog form BD and convert the band signal SIG BD into a digital signal SIG DIG For example, the ADC 230 can have a specification of 30 megasamples per second (MSPS). The ADC 230 with a sampling rate of 30 MSPS can extract 30 million signal samples per second. According to the Nyquist sampling theory, the sampling process performed at a sampling rate of 30 MSPS can be sufficient to capture signals having a frequency above 100 kHz and below 5 MHz. The ADC 230 can sample the band signal SIG BD and extract digital values indicating the amplitude values of the analog signal at specific time points. The ADC 230 can send the digital signal SIG DIG to the DSP 240.

[0067] According to some embodiments, the signal processor 200 may further include a digital oscillator, a digital mixer, and a band-pass filter. For example, the digital oscillator, the digital mixer, and the band-pass filter can be located between the ADC 230 and the DSP 240.

[0068] The digital oscillator and the digital mixer can convert the frequency of the digital signal SIG DIG The band-pass filter can be a filter that only allows signals between specific frequencies to pass through. The band-pass filter can receive the digital signal SIG from the ADC 230DIG extract only the signals between specific frequencies. For example, a band-pass filter can be a low-pass filter. However, the above description is only an example and does not limit the embodiments according to the present disclosure.

[0069] The DSP 240 can process the digital signal SIG DIG perform a Fourier transform (FT) to transform the digital signal SIG DIG into a frequency-domain signal SIG FD . The DSP 240 can extract the frequency-domain signal SIG DIG by decomposing the digital signal SIG in the time domain FD into frequency components. The DSP 240 can generate multiple signal strength values SS from the frequency-domain signal SIG FD . The multiple signal strength values SS can include characteristics (such as noise, etc.) of the RF signal SIG RF . The DSP 240 can perform signal processing on the digital signal SIG DIG by using various types of algorithms. For example, the DSP 240 can extract the frequency-domain signal SIG DIG by performing a fast Fourier transform (FFT) on the digital signal SIG in the time domain FD . The DSP 240 can send the multiple signal strength values SS to the signal processor controller 250.

[0070] The signal processor controller 250 can control the antenna 210, the RF tuner 220, the ADC 230, and the DSP 240 in response to a trigger signal SIG TRG . For example, the signal processor controller 250 can be an SDR controller.

[0071] The signal processor controller 250 can send the multiple signal strength values SS to the contactor state estimator 300. The signal processor controller 250 can communicate with the user interface. The signal processor controller 250 can communicate with an external network and hardware. For example, the signal processor controller 250 can generate RF raw signal data including a noise pattern based on the received RF signal SIG RF and send the RF raw signal data to the contactor state estimator 300.

[0072] The contactor state estimator 300 can generate a state estimation value SEV for estimating the state of the contactor 11 by receiving the multiple signal strength values SS and inputting the multiple signal strength values SS into a pre-trained neural network model. The contactor state estimator 300 can analyze the noise data by using an algorithm and estimate the RF signal SIG RFThe characteristics, causes, impacts, etc. of the noise included. For example, the noise data may include information such as the average amplitude of the signal and the peak frequency of the noise.

[0073] The contactor state estimator 300 can detect the abnormal state of the contactor 11 by using the noise data extracted from the pre-trained ANN learning from the RF signal SIG RF The contactor state estimator 300 can learn the state data of the contactor 11 by using the ANN, and determine whether the state of the contactor 11 is a normal state or an abnormal state based on new data (e.g., the RF signal SIG RF ) generated from the contactor 11. For example, the ANN can be a convolutional neural network (CNN). The CNN can extract useful features from the frequency-based noise pattern, and on this basis, extract the operating state of the contactor 11.

[0074] According to some embodiments, the contactor state estimator 300 can send the state estimate value SEV of the contactor 11 to the BMS1000. For example, the contactor state estimator 300 can send the state estimate value SEV of the contactor 11 to the BMS1000 through the SPI protocol. For example, the state estimate value SEV of the contactor 11 can be an analysis data packet in the form of noise pattern: [data], frequency: [data], amplitude: [data].

[0075] Figure 3 is a schematic block diagram of a computing device for performing the method of operating the BMS according to some embodiments of the present disclosure.

[0076] Reference Figure 3 According to some embodiments of the present disclosure, the computing device 10 may include a memory 20 and a processor 30.

[0077] The memory 20 may be a record medium readable by the computing device 10, and may include permanent mass storage devices such as random access memory (RAM), read only memory (ROM), and disk drives.

[0078] The memory 20 may perform the function of temporarily or permanently storing the data processed by the processor 30. The memory 20 may include magnetic storage media or flash storage media, but the scope of the present disclosure is not limited thereto. For example, the memory 20 may receive multiple pieces of data constituting the ANN, and temporarily and / or permanently store the data. The memory 20 may store the training data for training the ANN. However, the above description is only an example, and the spirit and scope of the embodiments of the present disclosure are not limited thereto.

[0079] According to some embodiments of the present disclosure, the memory 20 may store program code for executing a method of operating the BMS, multiple pieces of data for executing the program code, and multiple pieces of data generated during the execution of the program code. The program code may include algorithm code for Fourier transform and ANN model operations.

[0080] According to the present disclosure, the memory 20 may store multiple pieces of data for executing a method of operating the BMS. For example, the memory 20 may store data for training the ANN. For example, the memory 20 may store multiple signal strength values generated by preprocessing the state of the contactor switched by the control signal of the battery manager and the RF signal generated during the switching of the state of the contactor as training data.

[0081] The processor 30 can generally control the overall operation of the computing device 10. The processor 30 may be configured to process instructions of a computer program by performing basic arithmetic, logical, and input / output operations. The processor 30 may receive training data and generate training result data from the received training data according to a preprocessing method selected by the user. For example, the processor 30 may receive RF signal data of the contactor and generate state data of the contactor from the RF signal data of the contactor.

[0082] The above-mentioned processor 30 may refer to, for example, a data processing device embedded in hardware, which has a circuit physically configured to execute functions represented by codes or instructions included in a program. Examples of data processing devices embedded in hardware may include processing devices such as microprocessors, central processing units (CPUs), processor cores, multi-processors, application-specific integrated circuits (ASICs), and field-programmable gate arrays (FPGAs), but the scope of the present disclosure is not limited thereto.

[0083] The processor 30 may receive data stored in the memory 20 and send the data to the memory 20. In some embodiments, according to some embodiments, in addition to the memory 20 and the processor 30, the computing device 10 may further include a communication module, input / output devices, storage devices, etc.

[0084] The operation of the processor 30 according to various embodiments is described in more detail below.

[0085] Figure 4 is a flowchart showing a method of operating the BMS according to some embodiments of the present disclosure. Although various operations are shown in Figure 4 , according to embodiments of the present disclosure, it is not limited thereto. For example, according to some embodiments, without departing from the spirit and scope of the embodiments of the present disclosure, the method of operating the BMS may include additional operations or fewer operations, and the order of operations may vary unless otherwise stated or implied.

[0086] Reference Figure 4 , the method of operating the BMS can be executed by the processor 30 of Figure 3 .

[0087] In operation S10, a control signal for controlling the state of a contactor connected to a battery can be output to the contactor. The control signal can include an on signal for switching the contactor to an open state and an off signal for switching the contactor to a closed state. In operation S10, a trigger signal can be extracted in response to the output of the control signal, and an RF signal can be captured in response to the trigger signal.

[0088] In operation S20, an RF signal generated during the switching of the state of the contactor can be received. In operation S20, an RF signal can be received from the contactor, and a band signal in a preset frequency band in the RF signal can be output. In some embodiments, the band signal can be transformed into a digital signal, a Fourier transform can be performed on the digital signal to transform the digital signal into a frequency-domain signal, and a plurality of signal strength values can be generated from the frequency-domain signal.

[0089] In operation S30, a plurality of signal strength values corresponding to a plurality of frequency bands can be generated based on the RF signal. In operation S30, a plurality of signal strength values can be generated in response to the trigger signal.

[0090] A state estimation value for estimating the state of the contactor can be generated by inputting the plurality of signal strength values into a pre-trained neural network model. The pre-trained neural network model can generate a state estimation value of the contactor according to the plurality of signal strength values received as input and the confidence score of the state estimation value. Reference Figures 5A to 5C The neural network model is described in more detail.

[0091] Operation S40 can further include an operation of detecting whether the contactor is in an abnormal state based on the state estimation value of the contactor and the confidence score of the state estimation value. For example, if the state of the contactor according to the state estimation value of the contactor in response to the on signal in operation S40 is a closed state, the state of the contactor can be determined to be an abnormal state. If in operation S40, the state of the contactor according to the state estimation value of the contactor in response to the off signal is an open state, the state of the contactor can be determined to be an abnormal state. If the state of the contactor according to the state estimation value of the contactor in operation S40 is a welded state, the state of the contactor can be determined to be an abnormal state.

[0092] Hereinafter, reference Figures 5A to 5C Aspects of an example structure of an ANN are described in more detail.

[0093] Figures 5A to 5C shows according to the present disclosure byFigure 3 View of an example structure of an ANN trained by the processor 30. Figures 5A to 5C The ANN model shown in can be included in Figure 1 the contactor state estimator 300 of.

[0094] Figure 5A It is a view showing an example structure of a pre-trained ANN model according to some embodiments of the present disclosure.

[0095] Referring to Figure 5A , the components of the ANN can include nodes / units, layers, weights, summation, and functions.

[0096] Nodes (or units) are components that make up each layer. The layer can be at least one of an input layer, a hidden layer, or an output layer. The input layer can be a layer that receives input data. The input layer can receive multiple input data (e.g., Figure 5A the x of 1 , x of 2 , …, x of p-1 and x of p ). p can be a natural number of 1 or greater. According to some embodiments, the multiple input data can be multiple RF signal data generated from contactors in the BMS. The hidden layer can process the multiple input data at least once. The hidden layer can be a layer that includes nodes (e.g., Figure 5A the and ) that multiply the multiple input data by weights and generate the result of the activation function. m can be a natural number of 1 or greater. The hidden layer can include multiple layers. According to some embodiments, for example, another hidden layer can be a layer that includes multiple nodes (e.g., and ). The output layer can be a layer that includes nodes (e.g., Figure 5A the o of 1 ,... and o of k ) that multiply the last hidden layer or the input layer by weights and generate the result value of the output function. k can be a natural number of 1 or greater. According to some embodiments, the result value of the output function can be the state data of the contactor generated based on the RF signal data generated from the contactor.

[0097] The function can be at least one of an activation function, an output function, or a loss function. The activation function can be a function that processes the sum of the product of a node and a weight and can calculate the node value of the hidden layer. The weight can indicate the connection strength between nodes. Summation can indicate summing the product of a weight and a node.

[0098] Referring to Figure 5A the example of, the k-th hidden node h of the l-th layer k (l)It can be defined according to the following formula: f (l) can be the activation function for the l-th hidden layer, w k (l) can be the weight vector, and b k (l) can be the bias for the k-th node of the (l + 1)-th layer. The output function can be a function that processes the product of the nodes and weights of the last hidden layer and can calculate the values of the nodes of the output layer (e.g., o 1 ,... and o k ). The loss function can be a function that measures the error between the result of the output function and the response value for learning the weights.

[0099] According to some embodiments, the above ANN model can be a neural network model pre-trained with training data including multiple signal strength values, and the multiple signal strength values are generated by preprocessing the state of a contactor switched by a control signal of a battery manager and an RF signal generated during the state switching of the contactor.

[0100] Figure 5B An example of the pre-trained ANN model according to some embodiments of the present disclosure is schematically shown.

[0101] The ANN according to some embodiments of the present disclosure can be an ANN according to a CNN model as shown in Figure 5B . In some embodiments, the CNN model can include multiple hidden layers and a classification layer. The CNN model can be a hierarchical model for finally extracting the features of the input data INPUT by alternately performing multiple operation layers (e.g., convolutional layer and pooling layer) in multiple hidden layers. In some embodiments, the Figure 3 processor 30 according to some embodiments of the present disclosure can establish or train the ANN model by processing the training data according to the supervised learning technique.

[0102] The Figure 3 processor 30 according to some embodiments of the present disclosure can generate a convolutional layer for extracting the feature values of the input data INPUT and a pooling layer for constructing a feature map by combining the extracted feature values. Feature information for classifying the input data INPUT can be extracted from the convolutional layer. Figure 3 The

[0103] According to some embodiments of the present disclosure Figure 3 Processor 30 of can generate a fully connected layer by combining feature maps, and the fully connected layer is arranged to determine the probability that the input data INPUT corresponds to each of a plurality of items. The fully connected layer can be a stage for determining classification in the final stage of the CNN model. The fully connected layer can include a fully connected layer that performs flattening to convert each layer into a one-dimensional vector and then connects the layers converted into one-dimensional vectors into one vector. The class with the highest probability can be classified as the output by using the softmax function. According to some embodiments of the present disclosure Figure 3 Processor 30 of can calculate an output layer including an output corresponding to the input data INPUT.

[0104] Figure 5C is an example schematically showing a process of training a pre-trained ANN model according to some embodiments of the present disclosure.

[0105] Figure 5C shows a CNN model, wherein Figure 2 Antenna 210 of SDR 200 receives an RF signal generated during the state of the switching contactor and uses the RF raw signal data as input data. For example, the RF raw signal data can include a noise signal. The RF raw signal data can include a pattern of the noise signal. The CNN model can be trained by using a plurality of signal strength values generated by preprocessing the state of the contactor and the RF signal generated during the state of the switching contactor as training data.

[0106] Figure 5C Conv1D of can indicate Figure 5B the convolutional layer of. The Conv1D layer can learn the features of the input data by using a filter. For example, the Conv1D layer can receive an RF signal as input data and learn the pattern of the RF signal as a feature of the input data. Figure 5C The ReLU activation function of can indicate Figure 5B the ReLU of. The ReLU activation function can add non-linearity to the input data. For example, the CNN model can learn a more complex pattern of the RF signal by adding non-linearity to the RF signal received as input data. Figure 5C The pooling layer of can indicate Figure 5B the pooling layer of. The pooling layer can downsample the data to prevent or reduce overfitting and improve computational efficiency. The process of the above Conv1D layer, ReLU activation function, and pooling layer can be repeated. The flattening layer and the fully connected layer are as described above with reference to Figure 5B stated.

[0107] The output data can be output data generated and classified based on the input data. According to some embodiments, the output data can be the status value of the contactor and the confidence score of the status value of the contactor. For example, according to the status value of the contactor, the status of the contactor can be any one of an open state, a closed state, and a welded state. The confidence score of the status value of the contactor can be a value greater than or equal to 0 and less than or equal to 1.

[0108] The ANN model can receive multiple pieces of RF signal data and send the estimated value of the status of the contactor (e.g., diagnostic data) to the BMS based on the data learned as described above.

[0109] Figure 5B and Figure 5C The ANN according to the CNN model shown in is an example, and the spirit of the present disclosure is not limited thereto. In some embodiments, the ANN can be stored in the above-mentioned memory in the form of coefficients of at least one node constituting the ANN, weights of the nodes, and coefficients of functions defining the relationships between multiple layers constituting the ANN. In some embodiments, the structure of the ANN can also be stored in the memory in the form of source code and / or programs.

[0110] The above reference Figures 5A to 5C The type and / or structure of the ANN described are examples, and the spirit of the present disclosure is not limited thereto. In some embodiments, various types of models of the ANN can correspond to the ANN described throughout the specification. According to some embodiments, models such as deep neural networks (DNN), CNN, recurrent neural networks (RNN), or bidirectional recurrent deep neural networks (BRDNN) can be used as the ANN according to some embodiments of the present disclosure, but the present disclosure is not limited thereto.

[0111] According to some embodiments of the present disclosure, compared with some methods of detecting the status of a contactor by using physical sensors, many hardware components may not be required, and the abnormal status of the contactor can be detected relatively more accurately by pre-learning the pattern of RF signals. In some embodiments, the safety and confidence level of the vehicle's BMS can be relatively improved by monitoring the status of the contactor in real time and detecting the abnormal status of the contactor in the BMS relatively quickly.

[0112] Some embodiments of the electronic or electrical devices and / or any other related devices or components according to the present invention described herein can be implemented using any suitable hardware, firmware (e.g., application specific integrated circuit), software, or a combination of software, firmware, and hardware. For example, the various components of these devices can be formed on one integrated circuit (IC) chip or on separate IC chips. Additionally, the various components of these devices can be implemented on a flexible printed circuit film, tape carrier package (TCP), printed circuit board (PCB), or formed on a substrate. Further, the various components of these devices can be processes or threads running on one or more processors in one or more computing devices, executing computer program instructions and interacting with other system components to perform the various functions described herein. The computer program instructions are stored in a memory, which can be implemented in a computing device using standard memory devices such as, for example, random access memory (RAM). The computer program instructions can also be stored in other non-transitory computer-readable media such as, for example, CD-ROM, flash drive, etc. Moreover, those skilled in the art should recognize that, without departing from the spirit and scope of the embodiments of the present invention, the functions of various computing devices can be combined or integrated into a single computing device, or the functions of a particular computing device can be distributed over one or more other computing devices.

[0113] The effects that can be obtained through the present disclosure are not limited to the above effects, and other technical effects not mentioned can be clearly understood by those of ordinary skill in the art from the description of the present disclosure described below.

[0114] The various embodiments of the present disclosure do not limit the scope of the present disclosure in any way. For the sake of simplicity of description, the description of some electronic components, control systems, software, and other functional aspects of the system can be omitted, unless it is necessary to enable those of ordinary skill in the art to understand, manufacture, and use the embodiments according to the present disclosure. The connection or connection unit of the lines between the components shown in the drawings can indicate a functional connection and / or a physical or circuit connection, and can be implemented as various alternative or additional functional connections, physical connections, or circuit connections in an actual device. Unless there are specific annotations such as "necessary" or "significantly", they may not be necessary components for implementing the present disclosure.

[0115] In the description of the present disclosure (especially in the claims), the use of the term "the" and similar indicative terms may correspond to both singular and plural forms. If a range is described in the present disclosure, it includes inventions that apply to each value within the range (unless there is a contrary description), and each individual value that constitutes the range is described in the description of the present disclosure. Unless there is an explicit order or statement contrary to the operations that constitute the method according to the present disclosure, the operations may be performed in any appropriate order. The present disclosure is not necessarily limited by the described order of operations. In the present disclosure, the use of all examples or exemplary terms (e.g., etc.) is merely for describing the present disclosure in detail, and unless limited by the claims, the scope of the present disclosure is not limited by the above examples or exemplary terms. Those of ordinary skill in the art will understand that various modifications, combinations, and changes can be made according to design conditions and factors within the scope of the appended claims or their equivalents.

[0116] Therefore, the spirit of the present disclosure should not be limited to the above embodiments, and not only to the claims described below, but all ranges that are equivalent or equivalently changed by these claims fall within the scope of the spirit of the present disclosure.

[0117] It should be understood that the embodiments described herein should be considered only in a descriptive sense and not for the purpose of limitation. The description of features or aspects within each embodiment should generally be considered applicable to other similar features or aspects in other embodiments. Although one or more embodiments have been described with reference to the accompanying drawings, those of ordinary skill in the art will understand that various changes in form and detail can be made therein without departing from the spirit and scope of the present disclosure as defined by the appended claims.

Claims

1. A battery management system, comprising: a contactor connected to the battery; a battery manager configured to output a control signal for switching a state of the contactor to the contactor; a signal processor configured to receive a radio frequency (RF) signal generated during the state switching of the contactor, and generate a plurality of signal strength values ​​corresponding to a plurality of frequency bands respectively based on the RF signal; as well as A contactor state estimator is configured to generate a state estimation value for estimating the state of the contactor by receiving the plurality of signal strength values ​​and inputting the plurality of signal strength values ​​into a pre-trained neural network model.

2. The battery management system according to claim 1, wherein: The battery manager is further configured to output a trigger signal to the signal processor in response to the output of the control signal, and the signal processor is further configured to capture the RF signal in response to the trigger signal.

3. The battery management system according to claim 2, wherein: The signal processor comprises: an antenna configured to receive the RF signal; an RF tuner configured to output a frequency band signal in a preset frequency band from the RF signal; an analog-to-digital converter ADC, configured to convert the frequency band signal into a digital signal; a digital signal processor DSP configured to transform the digital signal into a frequency domain signal by performing Fourier transform on the digital signal, and generate the plurality of signal strength values ​​from the frequency domain signal; and A signal processor controller is configured to control the antenna, the RF tuner, the analog-to-digital converter, and the digital signal processor in response to the trigger signal and send the plurality of signal strength values ​​to the contactor state estimator.

4. The battery management system according to claim 3, wherein: The preset frequency band ranges from 100 kHz to 5 MHz.

5. The battery management system according to claim 1, wherein: Widths of the plurality of frequency bands are equal to one another on a logarithmic scale.

6. The battery management system according to claim 1, wherein: Widths of the plurality of frequency bands are equal to each other.

7. The battery management system according to claim 1, wherein: The neural network model is a convolutional neural network (CNN) model and is pre-trained using training data including the multiple signal strength values, which are generated by pre-processing the state of the contactor switched by the control signal of the battery manager and the RF signal generated during switching the state of the contactor.

8. The battery management system according to claim 1, wherein: The control signal includes: an on signal for switching the contactor to an open state; and an off signal for switching the contactor to a closed state.

9. The battery management system according to claim 8, wherein: The pre-trained neural network model is configured to output the state estimate value based on the multiple signal strength values ​​and the confidence score of the state estimate value received as input.

10. The battery management system according to claim 9, wherein: The battery manager is configured to detect whether the contactor is in an abnormal state based on the state estimation value and a confidence score of the state estimation value.

11. The battery management system according to claim 10, wherein: The battery manager is also configured to determine the state of the contactor as the abnormal state based on at least one of the closed state according to the state estimation value of the contactor in response to the connection signal, the disconnected state according to the state estimation value of the contactor in response to the disconnection signal, and the welding state according to the state estimation value of the contactor.

12. A method of operating a battery management system including a contactor connected to a battery, the method comprising: Outputting a control signal for switching the state of the contactor to the contactor; receiving a radio frequency (RF) signal generated during switching of the state of the contactor; generating a plurality of signal strength values ​​corresponding to a plurality of frequency bands, respectively, based on the RF signal; as well as A state estimation value for estimating the state of the contactor is generated by inputting the plurality of signal strength values ​​into a pre-trained neural network model.

13. The method according to claim 12, further comprising: extracting a trigger signal in response to the output of the control signal; as well as The RF signal is acquired in response to the trigger signal.

14. The method according to claim 13, further comprising: receiving the RF signal; outputting a frequency band signal in a preset frequency band from the RF signal; Converting the frequency band signal into a digital signal; transforming the digital signal into a frequency domain signal by performing a Fourier transform on the digital signal, and generating the plurality of signal strength values ​​from the frequency domain signal; and The plurality of signal strength values ​​are extracted in response to the trigger signal.

15. The method according to claim 12, wherein: The control signal includes an on signal for switching the contactor to an open state and an off signal for switching the contactor to a closed state.

16. The method according to claim 15, wherein: The pre-trained neural network model is configured to generate the state estimate value based on the plurality of signal strength values ​​and the confidence score of the state estimate value received as input.

17. The method according to claim 16, further comprising: Whether the contactor is in an abnormal state is detected based on the state estimation value and a confidence score of the state estimation value.

18. The method according to claim 17, further comprising: The state of the contactor is determined to be the abnormal state based on at least one of the closed state according to the state estimation value of the contactor in response to the closing signal, the open state according to the state estimation value of the contactor in response to the opening signal, and the welding state according to the state estimation value of the contactor.

19. A computer-readable storage medium having stored thereon a computer program for executing the method according to claims 12 to 18 by using a computing device.

Images